Assessment in learning - It's crucial
Dylan William focusing on the 'Assessment, learning and technology: prospects at the periphery of control'
The context is school-based education. Dylan set the scene by arguing that in this context quality control is more effective than quality assurance, as learning is not predictable based on inputs but needs to be modelled around outputs - so assessment is critical as it allows for quality control..
You can view the elluminate broadcast of the talk at any time after it is finished or live now.
Structure, Alignment., Governance and Technology all interact: "The future is further away that you think!!"
Technology example, White-boards - no impact on student achievement, as many schools made things worse as those who made it better.
Three generations of school effectiveness research
Raw Results - different schools, different results but taking calls out made the results hardly difference. Value added approaches yet variability of teachers is more critical than variability of Schools 6 months learning can be 2 years for the weakest teachers.
Australian data says teachers don't get to teach well until they are 6 years in and keep gong for the next twenty years. New teachers tend not to be good teachers - we need a love the one you are with strategy.
Folding arms context - Often to learn something you have to do it the wrong way to think about what you are doing.
Teachers do not create learning, they create the context for learning.
Dylan says we needed a pedagogy of engagement and contingency.
"Schools are places where pupils go to see teachers work" - "If teachers go home more tired than the learners the wrong people are doing the work"
Intelligence becomes more critical to people's jobs the older they get, people opt for cognitive nieces that appeal to them. The same happen in learning, the more students engage the more their 'intelligence' goes up - other kids forgo this option and exasperate their disengagement learning process.
Is motivation cause of effect? Dylan is arguing that the right level of challenge is necessary to engage pupils.
The largest impact on learning indicates that feedback makes a big difference. See the work of Nyquist (2003) it is not just feedback but activities that close the learning gap - feedback and the opportunity to inform the development. CF this Scottish view
Design assessment based on what students are attempting to do - Arnold, Steinberg and Mislevy
The question is not what is right or wrong, the learning is what has been misunderstood. Why did people get it wrong, what was the process, here the learning begins. Getting the questions right and exploring the answers are crucial. Don't get it right for the wrong reason - they get it right because they understand. Plausible distractors are so important.
What technologies can help this. We are now starting to see:
The big goal is to get automated unstructured feedback. Not enough to say where the students are, we need to provide enhancing feedback. Multidimensional models.
Systems are now becoming available, using hardware and software and less 'teachers bandwidth' - interpreted constructive response must be the vision for supporting learning.
The context is school-based education. Dylan set the scene by arguing that in this context quality control is more effective than quality assurance, as learning is not predictable based on inputs but needs to be modelled around outputs - so assessment is critical as it allows for quality control..
You can view the elluminate broadcast of the talk at any time after it is finished or live now.
Structure, Alignment., Governance and Technology all interact: "The future is further away that you think!!"
Technology example, White-boards - no impact on student achievement, as many schools made things worse as those who made it better.
Three generations of school effectiveness research
Raw Results - different schools, different results but taking calls out made the results hardly difference. Value added approaches yet variability of teachers is more critical than variability of Schools 6 months learning can be 2 years for the weakest teachers.
Australian data says teachers don't get to teach well until they are 6 years in and keep gong for the next twenty years. New teachers tend not to be good teachers - we need a love the one you are with strategy.
Folding arms context - Often to learn something you have to do it the wrong way to think about what you are doing.
Teachers do not create learning, they create the context for learning.
Dylan says we needed a pedagogy of engagement and contingency.
"Schools are places where pupils go to see teachers work" - "If teachers go home more tired than the learners the wrong people are doing the work"
Intelligence becomes more critical to people's jobs the older they get, people opt for cognitive nieces that appeal to them. The same happen in learning, the more students engage the more their 'intelligence' goes up - other kids forgo this option and exasperate their disengagement learning process.
Is motivation cause of effect? Dylan is arguing that the right level of challenge is necessary to engage pupils.
The largest impact on learning indicates that feedback makes a big difference. See the work of Nyquist (2003) it is not just feedback but activities that close the learning gap - feedback and the opportunity to inform the development. CF this Scottish view
Design assessment based on what students are attempting to do - Arnold, Steinberg and Mislevy
The question is not what is right or wrong, the learning is what has been misunderstood. Why did people get it wrong, what was the process, here the learning begins. Getting the questions right and exploring the answers are crucial. Don't get it right for the wrong reason - they get it right because they understand. Plausible distractors are so important.
What technologies can help this. We are now starting to see:
- Automated essay scoring
- paraphrase analysers
- graphical analyses
The big goal is to get automated unstructured feedback. Not enough to say where the students are, we need to provide enhancing feedback. Multidimensional models.
Systems are now becoming available, using hardware and software and less 'teachers bandwidth' - interpreted constructive response must be the vision for supporting learning.
Labels: ALT-C2007, assessment, feedback, schools
posted by Anonymous at 11:06 am
3 Comments:
Interestingly Dylan was quite negative on personalised learning when Derek Morrison asked him about it. It was after I'd blogged, but this comment will remind me about it.
negative about personlised learning? but from his prsentation I won't have thought so? What did he say, Haydn?
When asked where he stood on personalised learning, he suggested it was two expensive and not necessarily what students needed. He said timely feedback which takes the learner forward might feel personal, but would not actually be personalised.
So I guess the question is the definition of personalised that you are using.
Post a Comment
<< Home